shadcn-ui

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs fetching component code from public registries (e.g., "npx shadcn@latest add " and "copy component code from ui.shadcn.com") and configuring custom registries/components.json with external URLs (e.g., registryDependencies including https://example.com/... or @ai/chatbot-rules), which requires the agent to ingest public third‑party registry content that can change installed code and drive subsequent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill includes runtime install commands (e.g., npx shadcn@latest add ...) that fetch and execute remote component code from the shadcn registry (e.g., https://ui.shadcn.com), so the registry URL is a runtime external dependency that can execute remote code and materially controls the installed components.